HIV Prevention With Positives: Often Ignored, Yet Often Effective

Table of Contents

• Introduction

• Prevention Counseling Takes Back Seat in Many HIV Clinics

• Brief Provider Prevention Prompting Has Big Impact

• A Faster, Simpler Approach to Prevention With Positives

• Why Providers Skip HIV Prevention Counseling So Often

• How Often Does HIV Get Transmitted?

• Are US HIV Transmission Rates Changing?

• HIV Transmitter Candidates Not Hard to Find

• References

Introduction

Everyone who gets HIV gets it from someone else. In the United States HIV usually jumps from one person to another during sex, and less often with needle sharing. Antiretrovirals for pregnant women and blood screening stoutly block the virus from infecting newborns and transfusion recipients. Even in countries with limited perinatal antiretroviral programs, sexual HIV transmission accounts for the bulk of new infections.

So if everyone with HIV stopped having unsafe sex and sharing tainted needles, the epidemic would end.

Yes, pregnant women with undiagnosed HIV or living in places without ready antiretroviral prophylaxis would pass HIV to offspring. But there would be fewer and fewer HIV-burdened pregnant women if HIV-positive men and drug-injecting partners stopped infecting them. So if sexual and drug-sharing transmission can be curbed, HIV infection would not be epidemic. Of course helping HIV-positive people in care stop transmitting their virus would do nothing to stymie transmission from people who don't know they're infected. But it would make a big dent in HIV incidence, and a bigger dent if wider HIV testing succeeds.

You might think, then, that HIV clinicians -- a group probably more attuned than any other to the public health consequences of the disease they treat -- would make preventing HIV transmission a top priority of day-to-day care. But study after study shows many don't -- even while a raft of other research proves that simple, regular counseling by HIV providers defuses risky behavior. And innovative risk-reduction and screening tools patients can self-administer promise to save clinician time without costing a bundle.

"Positive prevention" -- helping HIV-positive people avoid passing their virus to sex mates and injecting allies -- clambered several rungs up the US public health ladder in 2003 when the CDC promulgated its guidelines on Incorporating HIV Prevention Into the Medical Care of Persons Living With HIV.¹ Until then, CDC experts noted, "HIV prevention in this country ... largely focused on persons who are not HIV infected, to help them avoid becoming infected."¹ But to many that oak-to-acorn inversion has the logic of counseling potential car-crash victims to watch for reckless drivers rather than getting reckless drivers to slow down and stay in lane. And focusing on people without HIV had done nothing to stunt HIV incidence -- the new infection rate -- in the United States. (The CDC will update its positive-prevention guidelines in 2012. See the interview with Kathleen Irwin in this issue of RITA!)

If positive prevention languished as a clinical priority for nearly a decade after the CDC unveiled its guidance,¹ results of the randomized HPTN 052 trial pushed this epidemic-arresting strategy to a prime spot in the frontal lobes of HIV clinicians and everyone else who thinks about how the virus spreads.² Although this watershed study involved HIV-discordant couples, almost all of them heterosexual, few doubt that the foremost finding applies to anyone with HIV and that person's sex mates.

Randomizing HIV-positive partners in 1763 discordant couples to start combination antiretroviral therapy (cART) at a CD4 count between 350 and 550 cells/mm³ or to wait for AIDS or a count of 250 cells/mm³, HPTN investigators determined that early treatment cleaved the risk of HIV transmission to the negative partner by a whopping 96% (hazard ratio [HR] 0.04, 95% confidence interval [CI] 0.01 to 0.27, P < 0.001).²

Here, dramatically, was proof that keeping viral loads in check with standard cART virtually voided chances of sexual HIV transmission.² The solitary person in the early-treatment group who transmitted HIV to his partner had just begun treatment. When RITA! asked top HIV clinician/researchers to name the one or two steps they recommend to keep HIV-positive people from sharing their virus, 16 of 18 said treat them with antiretrovirals, and 7 of those 16 specifically cited HPTN 052.

But "treatment as prevention" will not transform the transmission landscape overnight, even in countries with ready access to antiretrovirals. To blunt HIV incidence in a big way, providers will have to give positive prevention a top-drawer spot in their treatment plans, even if time is short and reimbursement beggarly. This issue of RITA! aims to help clinicians focus more sharply on positive prevention and to see what research says on which strategies work best. Although most of the research and guidelines reviewed involve HIV in the United States and countries with a similar epidemic, the major points largely apply to HIV clinics across the world. Interviews with two CDC officials and with Stephen Morin, a top positive-prevention investigator, analyze some of the most telling data. Quick summaries for clinicians appear through this issue, and patient handouts appear in and after the second review article.

Prevention Counseling Takes Back Seat in Many HIV Clinics

Studies that reckon how many HIV providers counsel positive patients on HIV prevention almost all report dreary results. The US Health Resources and Services Administration (HRSA) 2011 Guide for HIV/AIDS Clinical Care reports that "one third to three fourths of HIV medical providers do not ask their patients about sexual behavior or drug use."³

In 1998 and 1999 a CDC team asked 839 Californians with HIV (607 gay/bisexual men, 127 heterosexual men, and 105 women) in six public HIV clinics if a health worker ever talked to them about safer sex or HIV disclosure to partners.⁴ Almost one third (29%) claimed no physician, physician assistant, nurse practitioner, nurse, social worker, health educator, psychologist, or psychiatrist ever discussed safer sex, and 33% could not remember a physician ever mentioning safer sex (Figure 1). Only half of the study group said a health worker ever discussed HIV disclosure with them. About 45% of these people had visited their clinic more than 20 times.

Although the US HIV epidemic was dominated by homosexual transmission then and remains so today, multivariate analysis determined that gay/bisexual men were half as likely as heterosexual men to be counseled about safer sex (odds ratio [OR] 0.48, 95% CI 0.28 to 0.81).⁴ The CDC investigators suggested this finding "may indicate that some providers feel uncomfortable talking about homosexual behavior or that they may (mistakenly) assume that MSM [men who have sex with men] already know about the importance of prevention and thus do not need additional information."

This seeming squeamishness about discussing sex with gays also surfaced in a 2008 study of 3787 HIV-positive people in New York City, San Francisco, Los Angeles, and Milwaukee.⁵ Almost three quarters of this Healthy Living Project group (73%) were men, 50% were MSM, 49% were African American, 26% white, and 19% Hispanic. Median age stood at 41 years.

Only 1356 study participants -- a little over one third (36%) -- said they received HIV prevention services in the 3 months before their first Healthy Living Project interview.⁵ Compared with other study participants, MSM were twice as likely to report HIV transmission risk behavior (OR 2.04, 95% CI 1.46 to 2.85), but they were 31% less likely to receive HIV transmission risk services (OR 0.69, 95% CI 0.58 to 0.82). Compared with whites, blacks were almost twice as likely (OR 1.93, 95% CI 1.38 to 2.68) and Hispanics 50% more likely (OR 1.52, 95% CI 1.16 to 1.98) to receive such services. People currently taking antiretrovirals were 20% less likely to receive transmission services (OR 0.81, 95% CI 0.74 to 0.89).

Earlier research by this group involving 618 HIV-positive people in 16 publicly funded US clinics found that providers were less likely to counsel people about HIV prevention in the previous 6 months in clinics primarily serving MSM.⁶ This study also traced a correlation between provider belief that behavior change is unlikely in HIV-positive people and chances of HIV prevention counseling. The researchers call this pessimism "provider fatalism," though it will more likely prove fatal to HIV-positive people or their partners.

A CDC team assessed prevention counseling of positive people in a survey of 317 HIV physicians in Atlanta, Baltimore, Miami, and Los Angeles, 208 of them men and 109 women.⁷ Only 37% of these physicians reported always discussing HIV transmission risk with patients, though 84% always talked about antiretroviral adherence and 65% always discussed prophylaxis for opportunistic infections. Two thirds of these physicians (65%) strongly or somewhat agreed that they had enough time to provide the care and information their patients needed, but only 41% of that group always discussed HIV transmission risk.

Multivariate analysis determined that Hispanic physicians and Asian/Pacific islander physicians were more likely to discuss HIV transmission than other racial/ethnic groups (P = 0.03 and P = 0.0001), as were physicians who said they had enough time to care for patients (P = 0.003) and physicians who cared for fewer patients (P = 0.05). The study identified a trend toward more frequent HIV transmission counseling by female physicians (43.5% versus 34.1% of male physicians).

Dismal HIV prevention counseling rates also emerged from surveys of 618 HIV-positive people in Ryan White-funded clinics plus in-depth interviews with 16 clinic administrators, 32 primary care providers, 32 support service providers, and 64 patients.⁸ Half of the HIV group was black, one quarter white, and 19% Hispanic. Three quarters were men, 49% heterosexual, and 45% gay or lesbian. The largest proportion of HIV-positive people (48%) was between 35 and 44, while 30% were 45 or older.

Only 27% of 413 patients who had sex in the past 6 months reported having a general discussion about "safer sex and ways to prevent transmission to others" in that day's primary care visit, and only 56% did in the last 6 months.⁸ Only 7% reported discussing specific sexual activities that day, and only 27% did in the past 6 months. These people said their clinicians talked about HIV transmission significantly less than adherence to antiretrovirals (41.5%), emotional issues (34%), or diet and nutrition (33%). People attending clinics with set procedures for HIV prevention counseling were twice as likely to report such counseling (OR 2.17, 95% CI 1.41 to 3.32, P < 0.001).

Brief Provider Prevention Prompting Has Big Impact

This bad to middling record of provider prevention prompting is particularly vexing because other research offers ample evidence that HIV-positive people curb risky behavior when providers talk to them about it, even briefly. University of Southern California researchers tested the value of "brief safer-sex counseling" by providers at six California HIV clinics.⁹ In this study of 585 sexually active HIV-positive adults, two clinics stressed the positive consequences of safer sex, two stressed the negative consequences of unsafe sex, and two control clinics stressed antiretroviral adherence. About 85% of study participants were men, about 75% were MSM, about 40% white, and about 35% Hispanic.

Stressing the negative worked best in people with two or more sex partners or with casual partners when the study began, cutting chances of self-reported unprotected anal or vaginal intercourse almost 60% (OR 0.42, 95% CI 0.19 to 0.91, P = 0.03) when compared with the control group.⁹ This result was similar when the researchers considered only MSM. The negative approach had no measurable impact on risky sex among people with only one partner (who had low prestudy rates of unsheathed sex) or only a main partner. And stressing the benefits of safer sex did not trim chances of unprotected intercourse compared with the control arm. These researchers concluded that "brief provider-delivered safer-sex interventions are both feasible and effective at HIV clinics that serve a large number of patients."

HRSA sponsored a 5-year program to see whether HIV prevention strategies in 13 clinics cut transmission risk in positive people.¹⁰ Of the 3556 study participants, 70% were men, 64% were older than 40, 45% were heterosexual, 44% gay, 8% bisexual, 48% African American, 37% white, and 11% Hispanic. Three quarters of these people had at least one sex partner in the past 6 months, and they averaged 5 partners. One fifth of the study group (21%) reported unprotected anal or vaginal sex with an HIV-negative or HIV status-unknown partner in the last 6 months.

Researchers randomized these people to receive prevention counseling by medical providers and/or "prevention specialists" (a social worker or peer educator) or the standard of care.¹⁰ Compared with the standard-of-care group, people who received provider-delivered prevention interventions had a 45% lower sexual transmission risk rate after 12 months (OR 0.55, 95% CI 0.32 to 0.94, P < 0.03). People randomized to counseling by a prevention specialist had a 42% lower sexual transmission risk rate at 6 months (OR 0.58, 95% CI 0.35 to 0.96, P < 0.04), but that benefit lost statistical significance at 12 months (OR 0.67, 95% CI 0.39 to 1.14, P < 0.14). People counseled on prevention by providers and prevention specialists did not have a significantly lower sexual transmission risk at 6 or 12 months than the control group.

The 12-month difference in transmission risk between provider groups and transmission-specialist groups probably reflects the ongoing transmission counseling offered by providers versus specialist counseling delivered only in the days or weeks after initial patient interviews. These researchers concluded that "behavioral interventions are most effective if they are delivered in 'doses' -- such as at routine medical care visits -- over time."¹⁰

There's another advantage to the provider-only approach: It's the cheapest way to go.¹¹ After 3 years of the HIV prevention demonstration project just described,¹⁰ the investigators figured costs and cost-effectiveness for the three approaches used: clinical provider alone, prevention specialist (social worker or peer) alone, and provider-plus-prevention specialist. Cost per patient for 3 years averaged $1004 when only the clinical provider worked with the patient on prevention, $3173 when only the prevention specialist worked with the patient, and $3430 when both the clinical provider and the specialist got involved.¹¹ Compared with the lifetime cost of HIV/AIDS care and with other effective HIV prevention interventions, the clinical provider-led interventions proved cost-effective, but specialist-led interventions and provider/specialist-led interventions did not.

A 2001-2002 survey of 614 HIV-positive people interviewed immediately after an HIV clinic visit at one of 16 Ryan White-funded clinics in 9 states divided clinics into those with written HIV prevention procedures, those where providers offered prevention counseling on their own initiative, and those with no prevention procedures.¹² Half of the study group (51%) was black, 25% were white, and 19% were Latino. Almost three quarters of patients (73%) were men, 48% were heterosexual, and 46% were gay or bisexual. Three quarters (77.5%) were older than 35, three quarters (76.4%) were taking antiretrovirals, and two thirds (66.9%) had sex in the last 6 months.

How often these people got counseled about HIV transmission depended on their clinic's prevention policy: In clinics with written procedures, 69% said their provider talked to them about HIV transmission in the past 6 months, compared with 56% in clinics where individual providers had to take the initiative, and 45% in clinics with no HIV prevention protocol.¹² Compared with patients whose clinics had no prevention policy, those in clinics with spelled-out procedures had a tripled chance of getting prevention counseling (OR 3.17, 95% CI 1.24 to 8.06, P < 0.02). Notably, though, even in clinics with written guidance, almost one third of patients said their provider did not talk about prevention, and overall 43% said they had not discussed safer sex with their provider.

As in two studies reviewed above,⁴˒⁵ heterosexuals in this study were more likely to get prevention counseling in the past 6 months than gays and bisexuals (OR 1.47 in bivariate analysis, 95% CI 1.01 to 2.12, P = 0.042).¹² Blacks got counseled more than non-blacks (OR 1.64, 95% CI 1.09 to 2.46, P = 0.018), women more than men (OR 1.59, 95% CI 1.09 to 2.32, P = 0.016), and sexually active people more than those not active (OR 1.70, 95% CI 1.26 to 2.31, P = 0.001). People older than 35 proved less likely to hear their provider talk about prevention than younger people (OR 0.57, 95% CI 0.39 to 0.85, P = 0.005), as were people currently taking antiretrovirals (OR 0.57, 95% CI 0.38 to 0.86, P = 0.007). The only one of these factors that independently predicted prevention prompting in multivariate analysis in clinics with written prevention procedures was current antiretroviral therapy, which lowered chances of counseling more than 60% (OR 0.38, 95% CI 0.33 to 0.43, P = 0.001).

A Faster, Simpler Approach to Prevention With Positives

Despite the several just-reviewed studies showing the value of clinician-provided prevention counseling, at least one study identified a counselor who did better than clinicians in getting HIV-positive people to cut back on risky sex: a computer.¹³ This trial randomized 566 HIV-positive people in six Los Angeles clinics to (1) a 10-minute interactive computer program (two clinics), (2) a provider-delivered prevention interaction (two clinics), or (3) standard care (two clinics). Compared with people randomized to provider delivery or standard care, those who pointed and clicked their way through a prevention program reported a significant drop in number of HIV-negative or HIV status-unknown partners. And compared with the standard-care group, computer users claimed a significant drop in vaginal or anal sex without condoms.

Another interactive laptop prevention program, the Positive Steps "Video Doctor," yielded significant declines in illicit drug use, average days of ongoing drug use, unprotected sex, and number of casual sex partners compared with standard care.¹⁴ This randomized trial focused on 476 HIV-positive people who reported substance use or sexual risk when attending five HIV outpatient clinics in San Francisco. Results for the first two outcomes were statistically significant both 30 and 60 days after the intervention.

An advantage of both of these digital strategies is that they require 0 up-front provider time.¹³˒¹⁴ The Video Doctor gives the flesh-and-blood doctor a "cuing sheet" on risks reported by each patient so the provider can follow-up with appropriate questions and, if necessary, referral.

All three positive-prevention experts interviewed in this issue of RITA! stress the value of these time-saving approaches for busy HIV practices. And, no doubt, when it comes to intimate questions about sex and drugs, many people with HIV would rather respond to a computer than to a clinician. The table between the two interviews in this issue outlines these and other patient-administered prevention strategies and provides links for more information.

Why Providers Skip HIV Prevention Counseling So Often

If people with HIV respond to simple HIV prevention counseling from providers -- and waiting room laptops -- as part of their regular check-up, why do physicians and other practitioners skip this facet of HIV care so often?³⁻⁸ One HIV clinician/researcher told RITA! he could think of four reasons: "Time. Time. Time. Time." Another four reasons are money, money, money, and money. As the CDC delicately notes in its positive prevention guidelines, "Some clinicians have expressed concern that reimbursement is often not provided for prevention services and note that improving reimbursement for such services might enhance the adoption and implementation of these guidelines."¹

The survey of 317 HIV physicians in Atlanta, Baltimore, Miami, and Los Angeles underlined the importance of time in determining whether clinicians paused to stress positive prevention.⁷ Physicians who strongly agreed they had enough time "to provide care and information to patients" were over 3 times more likely to discuss prevention than physicians who said they didn't have enough time (OR 3.4, 95% CI 1.9 to 6.1, P < 0.001). Every additional 20 patients seen monthly lowered chances of always discussing transmission almost 10% (OR 0.92, 95% CI 0.84 to 0.99, P = 0.05).

Three studies reviewed above suggest another important reason -- provider discomfort in discussing sex with gay or bisexual men,⁴˒⁵˒¹² though research reviewed below indicates these men pose the biggest transmission risk among HIV-positive people in the United States.

Another already-mentioned study uncovered a nearly nihilistic reason why providers don't talk to HIV-positive people about prevention -- the belief that people with HIV won't change their behavior.⁶ Besides surveying 618 HIV-positive people about prevention counseling, these researchers conducted 144 interviews with providers, administrators, and patients in 16 publicly funded US clinics. Then they summed pessimistic comments (described as "provider fatalism") and assigned each clinic a "fatalism score."

People in high-fatalism clinics proved less likely to report prevention counseling than those in low-fatalism clinics, even after statistical adjustment for clinic characteristics, patients' sexual risk, and patients' health status. People in high-fatalism clinics were more likely to be white, gay, educated, and older -- a finding that also suggests providers in these clinics assume their patients are sophisticated or disciplined enough to avoid transmission risk on their own.⁶

A separate analysis of these patients found that people in their first year of care were twice as likely to hear from their provider about safer sex and transmission as people in care for more than 1 year (OR 2.35, 95% CI 1.60 to 3.44, P < 0.001).⁸ Interviews with these providers confirmed an array of familiar reasons for shunning prevention reminders: lack of time, specialized training, and dedicated funding for staffing. Interviews with some providers evinced a "conflict over their role and responsibility" when they described themselves "as advocates for their patients' health, rather than as guardians of the public health." Providers who agreed that HIV clinics should talk to patients about positive prevention did not agree on exactly who should do the talking.

Because research on this question is lean, RITA! asked an impromptu panel of clinician/researchers to offer their insights on why providers don't spend more time stressing positive prevention (see box). Almost all 18 ticked off the same reasons disclosed in studies outlined above -- the (flattering but perhaps unfounded) assumption that patients are avoiding risk without coaching, the belief that primary care doesn't carry a public health mandate, and too little time:

"My patients often come with lists of issues to deal with and we attend to those first, often leaving little time for issues that would be on my list," explained Ann Collier. "Other issues of care often seem more important, more pressing, or more rewarding," David Margolis noted, acknowledging that "this is a problem." Indeed, too little time may soon become a time-worn excuse for shunning prevention counseling. The forthcoming CDC update on prevention with positives will stress interventions that require little or no clinician time, as discussed in the two interviews in this issue, in the table between those interviews, and in the second review article.

Providers probably also look beyond the time taken for prevention counseling itself to what might follow: If a tricky concern does arise during counseling, clinicians may fear they will have little time to deal with it meaningfully. Along the same lines, John A. Bartlett suggested that hearing patients say they haven't backed away from risky sex creates a "burden" for the provider, "and we might wish that we didn't have this information." But that fear, he stressed, does not excuse clinicians from talking about risky sex.

Carl Fichtenbaum underlined the futility some clinician feel in getting patients to avoid transmission-risk behaviors, the "provider fatalism" seen in the 16-center study discussed above.⁶

David Wohl emphasized provider discomfort in talking explicitly about sex. "Even HIV providers used to doing anal paps and hearing about sex toys and meth-fueled nights on the town can get shy when it comes to talking dirty," he offered. "We are also busy and sex talk takes time."

Some providers, Ian Frank suggested, fear that hammering on consistent condom use may send a pejorative message to the patient, as if the clinician is saying, "You are doing something you know you shouldn't be doing!" He sees value in "getting beyond the condom conversation, which is important, but not the only way to counsel patients about transmission risk."

Paul Sax also stressed that providers must not seem judgmental: "For newly diagnosed patients, or patients seeing you for the first time, it's critically important not to imply blame for the patient being HIV positive," he explained. "Consciously or subconsciously, we may feel that discussing prevention sends that message -- that if only you had followed this prevention advice, you wouldn't be infected."

How Often Does HIV Get Transmitted?

TB may jump from one person to the next more readily than HIV makes that leap, but odds of blood-borne or sexual transmission of HIV are not long when one considers (1) the consequences and (2) how often people embrace the risk in a lifetime. Getting transfused with HIV-tainted blood -- though rare today -- almost always results in HIV infection, with chances estimated at 95-in-100.³ Children born to HIV-positive mothers without antiretroviral prophylaxis stand a 1-in-4 chance of HIV infection. From there, odds drop in people sharing drug-injecting equipment or having condom-free sex (Figure 2).

But compared with other risks, sexual HIV transmission chances should inspire caution -- if not awe -- in people who let HIV near a nick, sore, or porous mucosal wall. For example, the 1-in-150 chance of picking up HIV while sharing injecting equipment exceeds the 1 in 167 risk of dying from heart disease or from an accident in the next 10 years among 45-year-old US men who never smoked¹⁵ and far exceeds the 1-in-1000 risk of dying from pneumonia in the next 10 years among 45-year-old US men who do smoke.¹⁵ The high estimate for acquiring HIV during rubber-free male-to-male anal intercourse, 1-in-10, is about the same as the 1-in-10 to 1-in-25 chance of miscarriage in the United States¹⁶ and far exceeds the 1-in-36 chance of rolling snake eyes (one and one) in craps.

Lifetime chances of HIV infection in the US are downright awful. For the years 2004-2005, CDC researchers estimated that white men in 33 of the United States ran a 1 in 104 chance of getting infected, Hispanic men a 1 in 35 chance, and black men a 1 in 16 chance.¹⁷ Estimated lifetime infection rates for US women were 1 in 588 for whites, 1 in 114 for Hispanics, and 1 in 30 for blacks. If you told people they had a 1 in 16 to 1 in 104 lifetime chance of dying in a plane crash, how many people would fly?

CDC number jugglers offer another way to weigh the risk of getting HIV from a sex partner (Figure 3).¹ Starting with the premise that insertive penile-oral sex is the least likely way to pick up the retrovirus, the CDC team uses study data to estimate the relative risk of receptive penile-oral sex (twice as likely), insertive vaginal sex (10 times more likely), receptive vaginal sex (20 times more likely), insertive anal sex (13 times more likely), and receptive anal sex (100 times more likely). Compared with using a condom, the CDC figures, shunning condoms hikes the risk of HIV acquisition 20 times.

Needle sharing and needlestick HIV transmission risk varies according to how much blood gets under the stuck person's skin. And anal or vaginal transmission risk depends on viral load in the already-infected person and sexually transmitted diseases or open wounds in either partner. A landmark study of 415 HIV-discordant Ugandan couples -- one partner positive, one negative -- established the tight link between HIV load in plasma and transmission risk in this antiretroviral-naive group.¹⁸ Viral load averaged 90,254 copies/mL in people whose partner got infected with HIV versus 38,029 copies/mL in those whose partners stayed HIV-free (P = 0.01). Multivariate analysis accounting for other transmission risk factors determined that every 10-fold higher viral load more than doubled transmission risk (OR 2.45, 95% CI 1.85 to 3.26).

A more recent analysis of 235 monogamous Ugandan couples in this same population confirmed the intimate link between transmission risk and viral load (as well as genitourinary disease) in the positive partner.¹⁹ These researchers examined the impact of disease stage, viral load, and other factors on transmission risk in the years before any of these people took antiretrovirals. They established HIV transmission by HIV sequence analysis.

The HIV transmission rate was about 8 per 1000 coital acts in the first 5 months of HIV infection, when viral load is highest (median 30,000 copies/mL in this cohort).¹⁹ Six to 35 months after infection, the rate fell to about 1.5 per 1000 coital acts as the viral load waned (median 2600 copies/mL 15 months after infection). During chronic HIV infection, transmission rates varied from about 0.5 to 1.0 per 1000 coital acts (median viral load 10,300 copies/mL at study entry and 15,000 copies/mL after 30 months of follow-up). In late-stage infection (median viral load 112,600 copies/mL), transmission rates climbed back above 1 per 1000 coital acts, reaching about 3.5 per 1000 in the 6 to 25 months before death.

Multivariate analysis determined that, compared with a viral load in the lowest quartile, a load in the next-higher quartile tripled the risk of HIV transmission per coital act (rate ratio [RR] 3.31, 95% CI 1.01 to 10.80), a load in the next-higher quartile upped the risk more than 6 times (RR 6.39, 95% CI 2.10 to 19.42), and a load in the highest quartile inflated odds 7 times (RR 7.06, 95% CI 2.29 to 21.81). In a model that also included viral load, genitourinary disease in the initially infected partner doubled the risk of transmission (RR 2.05, 95% CI 1.02 to 4.14).

A recent longitudinal study of 1381 initially HIV-negative gay men in Sydney, Australia figured a per-sex-act HIV risk of 14.3-in-1000 for men who practiced unprotected receptive anal intercourse with ejaculation in the rectum versus 6.5-in-1000 for unprotected receptive anal intercourse with withdrawal before ejaculation.²⁰ The latter rate came close to the 6.2-in-1000 for uncircumcised men who practiced unprotected insertive anal intercourse, much higher than the 1-in-1000 rate for circumcised men taking the insertive role. The 1.43-in-100 infection rate for receptive partners whose mates ejaculated in their rectum looks big when one considers that the 663 men who took that role reported 56,514 such encounters in 6 years of total follow-up, or 85 per man. Also, these findings underline the iffy protective value of withdrawal during anal sex.

Are US HIV Transmission Rates Changing?

Thanks to busy CDC number summers, the United States has well-estimated and routinely updated counts of HIV prevalence, incidence, and (less often) transmission. But discerning precisely what these tallies say about positive prevention takes an extra army of experts.

In 2008 the CDC heralded a 1977-2006 estimate of HIV transmission with the headline "Dramatic Declines [in HIV transmission] Indicate Success in U.S. HIV Prevention."²¹ The CDC summary explained work by Johns Hopkins researchers who used CDC data to reckon HIV infections transmitted per 100 HIV-positive people by dividing HIV incidence (the number of new infections) for a given year by HIV prevalence (the overall number of people with HIV) for the same year.²²

Three years after clinicians identified AIDS, in 1984, the Hopkins team figured a transmission rate of 44.4%, meaning 44 out of every 100 HIV-positive people passed the virus to someone else.²² That rate plunged to 11.7% in 1990, to 7.5% in 1997 (a year or 2 after triple therapy arrived), and to 5.0% in 2006, the final year of the analysis (Figure 4). The nosedive from 44 to 5 transmissions per 100 positive people represents an 89% decline. So something -- probably many things -- were going right.

The Hopkins formula for calculating transmission rate (incidence/prevalence in a given year²²) means transmission would drop if incidence fell and/or if prevalence rose. Indeed, US HIV incidence did tumble from its estimated peak of 130,400 new infections in 1984 and 1985 to about 84,000 from 1986 to 1990, then to about 48,000 in 1991 to 1996.²² From 1997 through 2006, the last year of this analysis, HIV incidence fell no farther.

Over the same years, HIV prevalence climbed steadfastly from about 400,000 cases in 1985, to about 800,000 in 1998, and up to 1.2 million in 2008, the last year for which the CDC made an estimate.²³ These estimates include people with undiagnosed HIV. US HIV prevalence rose so doggedly after combination therapy arrived not because lots more people were getting infected (incidence was flat), but because already infected people were living longer. Much longer. Together, falling (or flat) incidence plus rising prevalence would make the transmission rate (incidence/prevalence) sink.

Despite the dwindling estimated transmission rate through 2006, HIV incidence among people 13 and older bottomed out that same year at around 48,000 new infections per year, according to a 2011 CDC study.²⁴ From 2006 through 2009, incidence stayed stuck at that level -- 48,600 new infections in 2006, 56,000 in 2007, 47,800 in 2008, and 48,100 in 2009. But because HIV prevalence continued to climb, at least through 2008,²³ the CDC's Kathleen Irwin believes transmission rates were falling in those years (see the interview in this issue): If there are more and more people with HIV from one year to the next but incidence (new infections) stays flat, the transmission rate must be dropping.

What these numbers don't explain is why HIV transmission rates appeared to keep dipping in the 15 years since triple therapy arrived. Almost certainly, pushing viral loads to uncountable levels in more and more people stanched transmission considerably.²⁵⁻²⁷ Whether prevention counseling or risk reduction has much to do with waning transmission seems dubious, since these studies also found rising rates of sans-condom receptive anal intercourse,²⁵ vaulting prevalence of infectious syphilis, genital gonorrhea, and genital chlamydia,²⁶ and climbing rates of rectal gonorrhea.²⁷

The CDC incidence study had no trouble pinpointing groups whose new HIV infection rate marooned incidence on a high plateau:²⁴ "The only population with a change in HIV incidence over the entire four-year [2006-2009] period was 13-29 year olds, and within that age group, the only risk group experiencing increases was MSM," the CDC reported. "Among young MSM the estimated number of new infections increased significantly from 2006 to 2009; the increase in incidence in this group was largely driven by a statistically significant increase in new HIV infections of 48% (12.2% annually) in young, black/African American MSM."²⁴

HIV Transmitter Candidates Not Hard to Find

HIV-positive people unaware of their viral stowaways constitute a deep and covert pool of potential transmitters. In 2006 the CDC figured that HIV-positive but undiagnosed people have a 3.5 times higher risk of passing HIV to a sex partner than people who know they carry the virus.²⁸ Wider HIV screening and earlier diagnosis can curtail HIV transmission, and clinicians can promote HIV testing through their positive patients and in their community -- topics examined in the preceding issue of RITA!.²⁹ But this review focuses on how providers can limit transmission from people who know they have HIV.

The list of likely transmitters is no secret to HIV providers:

• People with a high viral load

• Sexually active gay and bisexual men (MSM)

• Heavy injection and noninjection drug users

• People with mental health problems

But some studies identify transmission-risk groups one might not readily suspect.

High Viral Load and Recent HIV Infection

Ugandan studies of heterosexual couples (discussed above under "How often does HIV get transmitted?") framed a tight link between viral load and sexual transmission risk.¹⁸˒¹⁹ Almost a decade earlier, US researchers found sky-high levels of infectious virus (up to 1000 tissue culture-infective doses per milliliter of plasma) 6 to 15 days after the onset of primary infection symptoms.³⁰ Those loads "fell precipitously" by day 27.

Modeling blood and semen HIV loads in 30 men with known dates of infection or onset of acute HIV symptoms, another US team estimated that -- depending on how often they have sex -- men with average seminal loads during primary infection would infect 7% to 24% of female sex partners during the first 2 months of infection.³¹ If either partner has a sexually transmitted disease, those proportions would grow. In Quebec phylogenetic analysis of HIV from 591 people with primary infection and 795 with chronic infection yielded numbers suggesting that early

HIV infection accounted for half of HIV transmissions in the whole group.³² HIV providers understand the higher transmission risk conferred by lofty viral loads, but many people with early infection don't. In-depth interviews with 34 newly infected people in 6 US cities showed that "most participants knew little about the meaning and/or consequences of acute HIV infection, particularly that it is a period of elevated infectiousness."³³ Health workers who see acutely or recently infected adults should make sure they understand the link between high viral load and HIV transmission.

Gay and Bisexual Men

The CDC's latest HIV incidence study left no doubt that MSM still comprise the premier risk group in the United States.²⁴ HIV incidence among all people 13 and older remained flat at 48,000 new infections yearly from 2006 through 2009, but the new case rate did grow in one group: boys and men between 13 and 29 years old. Among young blacks, HIV incidence rose 12% annually over those 4 years (Figure 5). MSM accounted for 61% of new HIV infections in 2009, and MSM who injected drugs made up another 3%. A versatile few of these newly infected gays and bisexuals may be picking up HIV from women, but most are netting the retrovirus from other MSM.

Studies of HIV-positive people in primary care bear out this torqued-up transmission risk in MSM. In a 15-city US study, 2109 HIV-positive MSM, 1104 women, and 803 men who have sex with women (MSW) completed a computer-assisted survey asking how often they had unprotected anal or vaginal sex with an HIV-negative partner or a partner with an unknown HIV status.³⁴ Compared with MSW, MSM had more than a doubled risk of condom-free sex (OR 2.35, 95% CI 1.84 to 3.00, P < 0.001).

In a result that may surprise some, women in this study reported more than a 50% higher rate of risky sex than MSW (OR 1.56, 95% CI 1.19 to 2.05, P < 0.001).³⁴ An earlier study of 3723 HIV-positive people in Los Angeles, Milwaukee, New York City, and San Francisco found that 19% of women, 15.6% of MSM, and 13.1% of MSW had unprotected vaginal or anal intercourse with HIV-negative or status-unknown partners.³⁵ But this study of 1918 MSM, 978 women, and 827 MSW found that fewer than one quarter of women and MSW had 2 or more sex partners, compared with 59% of MSM. Researchers estimated that 30 sex partners of these HIV-positive people would get infected during the 3-month study, and 24 of them (80%) would pick up HIV from MSM.

Modeling HIV transmission data on the basis of partner information provided by 3652 MSM in five US cities, Emory University researchers figured that two thirds of transmissions in these men (68%) involve their main sex partner, not casual partners.³⁶ Sensitivity analyses plugging in different variables indicated that transmission from main sex partners may account for 52% to 74% of new HIV infections. Modeling also calculated that 69% of HIV infections resulted from receptive anal intercourse, 28% from insertive anal intercourse, and 2% from oral sex. Model-based estimated HIV incidence in this population was 2.2% per year.

The high transmission rate from main partners reflected more frequent sex with main partners, more frequent receptive anal intercourse with main partners, and lower condom use during anal sex with main partners.³⁶ These findings led the investigators to propose that "couples-based HIV prevention interventions for MSM should be given high priority in the US HIV prevention research portfolio." Not to mention the HIV clinical care portfolio.

Nearly everyone who follows the US HIV epidemic understands the high and abiding risk of viral transmission from gay and bisexual men. But as noted earlier in this article (see "Prevention Counseling Takes Back Seat"), that understanding has not spurred many HIV providers to talk to MSM frankly and frequently about protecting their partners from HIV.

Heavy Drug Use

Several studies confirm that using stimulants like crystal meth heightens the risk of HIV transmission. The already-discussed 15-city study of 4016 HIV-positive people in primary care found that stimulant use swelled transmission risk in gay men, straight men, and women (P < 0.05).³⁴

A four-city study of 1910 HIV-positive MSM (36% African American) who completed computer-assisted interviews about their 5 most recent partners traced significant links between three types of drug use and unprotected sex partners:³⁷ Stimulant use doubled the risk of unprotected sex with steady partners (adjusted OR 2.10, reference 1.00 to 4.39, P = 0.050), and both crystal meth (adjusted OR 1.76, reference 1.16 to 2.68, P = 0.009) and other drugs (adjusted OR 1.82, reference 1.16 to 2.86, P = 0.009) almost doubled the risk with casual partners. Neither education level nor race/ethnicity affected unprotected sex risk in these analyses. Men who did not disclose their HIV status to partners -- including steady partners -- had risky sex more often than serostatus disclosers.

A study of 90 HIV-positive MSM found that binge meth users reported more risky sexual behaviors -- as well as more social difficulties and physical and mental health problems -- than MSM who used meth but did not binge.³⁸ Forty-one of these 90 men (46%) said they binged for periods ranging from 2 to 33 days. Bingers were significantly more likely to be ethnic minorities and to have less education than non-bingers.

A study of 303 HIV-positive male and female African-American crack cocaine users defined binging as using as much crack cocaine as you can, until you run out of crack or can't use any more, in the last 30 days.³⁹ Half of these people (51%) reported binging for an average 3.7 days, during which they smoked 40 rocks of crack. Almost three quarters of bingers (72%) had sex during their last binge with an average of 3.1 partners. Recent male bingers were more likely than non-bingers to report lifetime and recent exchange of money for sex and drugs for sex. Recent female bingers were more likely to report lifetime trading of sex for drugs. Multivariate analysis determined that recent bingers were more likely than non-bingers to have high transmission risk scores, to have more sex partners in the last 1 month or 6 months, and never to use a condom during sex in the last 30 days.

Of course injection drug use poses a prodigious risk of HIV transmission, not only because HIV passes from person to person more readily through needle sharing than during sex (Figure 2), but also because of the social and psychological factors that boost HIV transmission risk among IDUs. And sharing injection works is not the only way IDUs transmit HIV, as indicated by a case-control comparison of 58 San Francisco IDUs who picked up HIV during follow-up and 1134 controls (matched for gender and date) who did not.⁴⁰ Compared with controls, the IDUs who became infected were almost 9 times more likely to be MSM (OR 8.8, 95% CI 3.7 to 20.5) and 5 times more likely to have traded sex for money in the past year (OR 5.1, 95% CI 1.9 to 13.7). Women with a steady sex partner who injected drugs were almost 70% less likely to become infected (OR 0.32, 95% CI 0.11 to 0.92).

The CDC figures that drug injectors accounted for 36% of AIDS cases since the US epidemic began through 2000.⁴¹ The 2006-2009 CDC study of HIV incidence found that IDUs made up about 10% of new HIV cases in each of those years.²⁴

Mental Health and Other Factors

Changing mental health in HIV-positive people has complex interactions with HIV transmission risk, as illustrated in a 936-person longitudinal study of positive people enrolled in a transmission-prevention trial.⁴² This four-city study recruited HIV-positive people from community agencies, AIDS service organizations, and medical clinics for a trial of an individually administered cognitive-behavioral intervention; 851 people had at least one follow-up visit. All study participants had signs of severe neuropsychological impairment or psychosis and reported unprotected vaginal or anal intercourse in the past 3 months with a sex partner whose HIV serostatus was negative or unknown. Participants were randomized to start the Healthy Living Program cognitive-behavioral intervention immediately or after 25 months of follow-up.

Ages ranged from 19 to 66. Most participants (79%) were men, 46% were black, 32% white, and 14% Hispanic.⁴² Almost three quarters of these people (72%) reported having sex with other men, and 43% had less than a high school education. When the study began, 40% of participants had clinically significant depression, 36% had clinically significant anxiety, and 25% had both.

Upon study entry, lower levels of depressive symptoms in the entire study group and higher scores in the Positive States of Mind (PSOM) scale were associated with having more HIV-negative partners in bivariate analysis.⁴² In the intervention group, worse anxiety symptoms were related to lower numbers of HIV-negative partners at baseline, and lower PSOM scores were related to increasing numbers of HIV-negative partners over time. In the delayed-intervention group, increasing depressive symptoms were related to decreasing numbers of HIV-negative partners over time. For the entire study group, lower PSOM scores from assessment to assessment were "modestly" associated with more risky sex acts, and more anxiety symptoms were associated with more HIV-negative partners. Stephen Morin (University of California, San Francisco), one of the investigators, discusses this study in an interview in this issue of RITA!

A Women's Interagency HIV Study (WIHS) analysis found evidence that poor antiretroviral adherence ups the risk of low condom use in women with HIV.⁴³ This study involved 766 antiretroviral-treated women, 66% of whom reported at least 95% adherence. Among sexually active women, less than 95% adherence doubled the risk of inconsistent condom use (OR 2.17). In the entire study group, poor adherence independently doubled chances of active drug use (OR 2.27, 95% CI 1.39 to 3.33). The WIHS team believes their results "highlight the importance of discussing [sexual behavior and adherence] in relation to one another, particularly with regard to patients experiencing virologic failure."

Even being hungry may boost the risk of unsafe sex, at least among IDUs taking antiretrovirals. Researchers in British Columbia reached that conclusion in a longitudinal study of 420 IDUs enrolled in the AIDS Care Cohort to Evaluate Exposure to Survival Services from 2005 through 2009.⁴⁴ Almost three quarters of this group (71%) suffered from food insecurity, marked by hunger due to lack of access or means to acquire food. People with food insecurity were marginally less likely to have an undetectable viral load (31% versus 39%, P = 0.099). Multivariate analysis that controlled for age, gender, marriage or cohabitation, binge drug use, homelessness, and antiretroviral use determined that food insecurity almost tripled the risk of unprotected sex (adjusted OR 2.68, 95% CI 1.49 to 4.82).

The second review article in this issue of RITA! scrutinizes diverse strategies for helping HIV-positive people avoid transmitting their virus. One of the clinician/researchers we polled for their opinions on this question, Babafemi Taiwo (Northwestern University, Chicago), offered this admirably succinct advice: "1. Test as many people as possible. 2. Treat as many infected patients as possible." But the next article will ask readers in endure considerably more detail.

References

1. Centers for Disease Control and Prevention. Incorporating HIV prevention into the medical care of persons living with HIV: recommendations of CDC, the Health Resources and Services Administration, the National Institutes of Health, and the HIV Medicine Association of the Infectious Diseases Society of America. MMWR. 2003;52(No. RR-12). Accessed October 8, 2011.

2. Cohen MS, Chen YQ, McCauley M, et al. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med. 2011;365:493-505. Accessed October 10, 2011.

3. US Department of Health and Human Services Health Resources and Services Administration. Guide for HIV/AIDS Clinical Care. January 2011. Accessed October 10, 2011.

4. Marks G, Richardson JL, Crepaz N, et al. Are HIV care providers talking with patients about safer sex and disclosure? A multi-clinic assessment. AIDS. 2002;16:1953-1957.

5. Steward WT, Charlebois ED, Johnson MO, Remien RH, Goldstein RB, Wong FL, Morin SF. Receipt of prevention services among HIV-infected men who have sex with men. Am J Public Health. 2008;98:1011-1014.

6. Steward WT, Koester KA, Myers JJ, Morin SF. Provider fatalism reduces the likelihood of HIV prevention counseling in primary care settings. AIDS Behav. 2006;10:3-12.

7. Gardner LI, Metsch L, Strathdee SA, del Rio C, Mahoney P, Holmberg SD. Frequency of discussing HIV prevention and care topics with patients with HIV: influence of physician gender, race/ethnicity, and practice characteristics. Gend Med. 2008;5:259-269.

8. Morin SF, Koester KA, Steward WT, et al. Missed opportunities: prevention with HIV-infected patients in clinical care settings. J Acquir Immune Defic Syndr. 2004;36:960-966.

9. Richardson JL, Milam J, McCutchan A, et al. Effect of brief safer-sex counseling by medical providers to HIV-1 seropositive patients: a multi-clinic assessment. AIDS. 2004;18:1179-1186.

10. Myers JJ, Shade SB, Rose CD, et al. Interventions delivered in clinical settings are effective in reducing risk of HIV transmission among people living with HIV: results from the Health Resources and Services Administration (HRSA)'s Special Projects of National Significance initiative. AIDS Behav. 2010;14:483-492.

11. Marseille E, Shade SB, Myers J, Morin S. The cost-effectiveness of HIV prevention interventions for HIV-infected patients seen in clinical settings. J Acquir Immune Defic Syndr. 2011;56:e87-e94.

12. Myers JJ, Steward WT, Charlebois E, Koester KA, Maiorana A, Morin SF. Written clinic procedures enhance delivery of HIV "prevention with positives" counseling in primary health care settings. J Acquir Immune Defic Syndr. 2004;37(suppl 2):S95-S100.

13. Lightfoot M, Rotheram-Borus MJ, Comulada WS, Reddy VS, Duan N. Efficacy of brief interventions in clinical care settings for persons living with HIV. J Acquir Immune Defic Syndr. 2010;53:348-356.

14. Gilbert P, Ciccarone D, Gansky SA, et al. Interactive "video doctor" counseling reduces drug and sexual risk behaviors among HIV-positive patients in diverse outpatient settings. PLoS One. 2008;3(4):e1988. Accessed November 11, 2011.

15. Woloshin S, Schwartz LM, Welch HG. The risk of death by age, sex, and smoking status in the United States: putting health risks in context. J Natl Cancer Inst. 2008;100:845-853.

16. American Pregnancy Association. Miscarriage. Accessed October 14, 2011.

17. Hall HI, An Q, Hutchinson AB, Sansom S. Estimating the lifetime risk of a diagnosis of the HIV infection in 33 states, 2004-2005. J Acquir Immune Defic Syndr. 2008;49:294-297.

18. Quinn TC, Wawer MJ, Sewankambo N, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. N Engl J Med. 2000;342:921-929. Accessed October 20, 2011.

19. Wawer MJ, Gray RH, Sewankambo NK, et al. Rates of HIV-1 transmission per coital act, by stage of HIV-1 infection, in Rakai, Uganda. J Infect Dis. 2005;191:1403-1409.

20. Jin F, Jansson J, Law M, et al. Per-contact probability of HIV transmission in homosexual men in Sydney in the era of HAART. AIDS. 2010;24:907-913.

21. CDC HIV/AIDS Facts. HIV transmission rates in the United States. December 2008. Accessed October 14, 2011.

22. Holtgrave D, Hall HI, Rhodes PH, Wolitski R. Updated annual HIV transmission rates in the United States, 1977-2006. J Acquir Immune Defic Syndr. 2009;50:236-238. Search for Holtgrave and click on title. Accessed October 14, 2011.

23. Centers for Disease Control and Prevention. HIV Surveillance -- United States, 1981-2008. MMWR. 2011;60(21):689-693. Accessed November 10, 2011.

24. Prejean J, Song R, Hernandez A, et al. Estimated HIV incidence in the United States, 2006-2009. PLoS One. 2011;6(8):e17502.

25. Porco TC, Martin JN, Page-Shafer KA, et al. Decline in HIV infectivity following the introduction of highly active antiretroviral therapy. AIDS. 2004;18:81-88.

26. Montaner JS, Lima VD, Barrios R, et al. Association of highly active antiretroviral therapy coverage, population viral load, and yearly new HIV diagnoses in British Columbia, Canada: a population-based study. Lancet. 2010;376:532-539.

27. Das M, Chu PL, Santos GM, et al. Decreases in community viral load are accompanied by reductions in new HIV infections in San Francisco. PLoS One. 2010;5(6):e11068.

28. Marks G, Crepaz N, Janssen RS. Estimating sexual transmission of HIV from persons aware and unaware that they are infected with the virus in the USA. AIDS. 2006;26;10:1447-1450.

29. Mascolini M. Late HIV diagnosis: predictors, costs, consequences, and solutions. RITA!. Accessed October 19, 2011.

30. SJ Clark, MS Saag, WD Decker, et al. High titers of cytopathic virus in plasma of patients with symptomatic HIV-1 infection. N Engl J Med. 1991;324:954-960.

31. Pilcher CD, Tien HS, Eron JJ. Brief but efficient: acute HIV infection and the sexual transmission of HIV. J Infect Dis. 2004;189:1785-1792.

32. Brenner BG, Roger M, Routy JP et al. High rates of forward transmission events after acute/early HIV-1 infection. J Infect Dis. 2007;195:951-959.

33. Remien RH, Higgins JA, Correale J, et al. Lack of understanding of acute HIV infection among newly-infected persons-implications for prevention and public health: the NIMH Multisite Acute HIV Infection Study: II. AIDS Behav. 2009;13:1046-1053.

34. Morin SF, Myers JJ, Shade SB, Koester K, Maiorana A, Rose CD. Predicting HIV transmission risk among HIV-infected patients seen in clinical settings. AIDS Behav. 2007;11(suppl 5):S6-S16.

35. Weinhardt LS, Kelly JA, Brondino MJ, et al; National Institute of Mental Health Healthy Living Project Team. HIV transmission risk behavior among men and women living with HIV in 4 cities in the United States. J Acquir Immune Defic Syndr. 2004;36:1057-1066.

36. Sullivan PS, Salazar L, Buchbinder S, Sanchez TH. Estimating the proportion of HIV transmissions from main sex partners among men who have sex with men in five US cities. AIDS. 2009;23:1153-1162.

37. Morin S, Steward WT, Charlebois ED, et al. Predicting HIV transmission risk among HIV-infected men who have sex with men: findings from the Healthy Living Project. J Acquir Immune Defic Syndr. 2005;40:226-235.

38. Semple SJ, Patterson TL, Grant I. Binge use of methamphetamine among HIV-positive men who have sex with men: pilot data and HIV prevention implications. AIDS Educ Prev. 2003;15:133-147.

39. Harzke AJ, Williams ML, Bowen AM. Binge use of crack cocaine and sexual risk behaviors among African-American, HIV-positive users. AIDS Behav. 2009;13:1106-1118.

40. Kral AH, Bluthenthal RN, Lorvick J, Gee L, Bacchetti P, Edlin BR. Sexual transmission of HIV-1 among injection drug users in San Francisco, USA: risk-factor analysis. Lancet. 2001;357:1397-1401.

41. CDC. Drug-associated HIV transmission continues in the United States. May 2002. Last modified March 8, 2007. Accessed October 20, 2011.

42. Comulada WS, Rotheram-Borus MJ, Pequegnat W, et al. Relationships over time between mental health symptoms and transmission risk among persons living with HIV. Psychol Addict Behav. 2010;24:109-118. Accessed October 20, 2011.

43. Wilson TE, Barron Y, Cohen M, et al. The Women's Interagency HIV Study. Adherence to antiretroviral therapy and its association with sexual behavior in a national sample of women with human immunodeficiency virus. Clin Infect Dis. 2002;34:529-534.

44. Shannon K, Kerr T, Milloy MJ, et al. Severe food insecurity is associated with elevated unprotected sex among HIV-seropositive injection drug users independent of HAART use. AIDS. 2011;25: 2037-2042.